get the UIImage (or CGImage) from drawRect - iphone

Question

I'm calling the following code from drawRect
- (void) drawPartial:(UIImage *)img colour:(UIColor *)colour {
CGContextRef context = UIGraphicsGetCurrentContext();
CGContextSaveGState(context);
CGContextTranslateCTM(context, 0.0, self.bounds.size.height);
CGContextScaleCTM(context, 1.0, -1.0);
CGContextDrawImage(context, self.frame, img.CGImage);
if (colour!=nil) {
CGContextSetBlendMode (context, kCGBlendModeColor );
CGContextClipToMask(context, self.bounds, img.CGImage);
CGContextSetFillColor(context, CGColorGetComponents(colour.CGColor));
CGContextFillRect (context, self.bounds);
}
self.currentImage=UIGraphicsGetImageFromCurrentImageContext();
CGContextRestoreGState(context);
}
Infact I call it multiple times to build a composite image dynamically coloured in. Each img passed in is effectively a transparent overlay with a different part of the image on it.
I can build my compound image by calling this multiple times in drawrect. The issue comes when I want to update one part of the image. Ideally I would be able to call the above function and just change the one part: I've tried playing with self.clearsContextBeforeDrawing to no avail.
Then I thought I would try to keep a copy of the image from each draw, as a cached image of the state - that way I just need to overlay the new part on that: two calls rather than 15.
However the line self.currentImage=UIGraphicsGetImageFromCurrentImageContext() is not returning me the current image so I can't build a cached copy.
Any help on either approach would really be appreciated. (or obviously point me the way I should be doing this!)
EDIT
I also tried compositing the image separtly using virtually the same code, and then drawing that, but again I don't get the image out...
- (UIImage *) overlayImage:(UIImage *)srcImage withImage:(UIImage *)overlayImage ofColour:(UIColor *)colour {
CGSize size =srcImage.size;
CGRect box = CGRectMake(0, 0, size.width, size.height);
UIGraphicsBeginImageContext(size);
CGContextRef context=UIGraphicsGetCurrentContext();
CGContextTranslateCTM(context, 0.0, size.height);
CGContextScaleCTM(context, 1.0, -1.0);
CGContextDrawImage(context, box, srcImage.CGImage);
if (colour!=nil) {
CGContextSetBlendMode (context, kCGBlendModeColor ); //kCGBlendModeMultiply
CGContextClipToMask(context, box, overlayImage.CGImage); // respect alpha mask
CGContextSetFillColor(context, CGColorGetComponents(colour.CGColor));
CGContextFillRect (context, box);
}
UIImage *result = UIGraphicsGetImageFromCurrentImageContext();
//UIImage *result = [UIImage imageWithCGImage: CGBitmapContextCreateImage (context)];
UIGraphicsEndImageContext();
return result;
}

Answer 1

As a rule, you want -drawRect: to be as simple as possible.
Move your drawing code from -drawRect: into a method that creates and draws the entire image you want. Reduce -drawRect: to just compositing this image into the current context.

Answer 2

It's entirely possible that your UIImage object simply isn't retaining the stored data. Furthermore you should know that UIImage is not a mutable object type. You'll get memory leaks unless you release it before reassigning it. Try this:
UIImage *result = [UIGraphicsGetImageFromCurrentImageContext() retain];